Methods in Analyzing Exposure-Adjusted Incidence Rates in Phase III Clinical Trial: Case Study

• Activity Number: 674
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Serap Sankoh*+
• Companies: Millennium Pharmaceuticals, Inc.
• Keywords: Exposure Adjusted Incidence Rates; ; Confidence Intervals; ; Adverse Events ; Clincical Trials

Abstract

Clinical trials with long term follow up the crude percentage may not be appropriate because it does not account for the exposure time. In this presentation, two methods of calculating exposure adjusted incidence rates and confidence intervals will be illustrated using a phase III clinical data. In the first method, exposure adjusted incidence rate is defined as the number of patients experiencing the adverse events divided by the person-time at risk. Days of person time are defined as days from first dose date to study end date. In the second method, days of person time are defined as days from first dose date to adverse event (AE) onset date. For each AE, the person time was truncated after a patient experienced the AE, although the patient continued to contribute person-time for other AEs. We will discuss how two methods described above will affect the estimation and inference in exposure adjusted AE rates when exposure times vary.